Sep 24, 2013
No model survives first contact with real content
At Content Strategy Forum 2013 in Helsinki, in a great presentation on “Deblobbing in the Real World”, Jeff Eaton said something that resonated deeply with me:
No model survives contact with real content.
I whole-heartedly concur. But, something was still not right. A week later, all became clear. I needed to tweak the sentence, just a little, to make things right with the world again. So here goes:
No model survives first contact with real content.
Better. Definitely expand “first contact” to mean the first few encounters with real content, but it’s an important distinction to draw out for anyone designing structured content. Why you ask?
Models Evolve
No model is supposed to survive first contact with real content, but they need to survive future contact.
So go ahead and break your models, test them by running various scenarios using real content. Inspect, then adapt your models because that’s design. This kind of model design is best done early and often, rather than downstream in production where the cost of change is insane.
Modelling is an iterative and recursive design process. Like any big problem, break it down into smaller pieces, tackling manageable bite sized chunks within each iteration. For example, an IT company may start with product information, breaking it down into categories such as networks, software and hardware, addressing each area accordingly. Breaking up content to be designed within smaller iterations gives us the thinking space we need to deal with its inherent complexity. Modelling is also recursive. So from the identified high level concepts, we can progressively delve deeper, adding more detail to our content types as we learn more about the domain.
It’s important for the model to evolve using real content from the start. To test it; flex it; validate assumptions; manage expectations; and to ultimately communicate a shared vocabulary of the content. You can’t do that lorem ipsum.
Content Gembas
The closer the model is to reality, the greater its chances of survival in the real world.
In Lean, there is a Japanese term known as gemba. It means the real place. The gemba is where your customers request and receive value from you. So in order to learn what your customers truly value, we need to spend time at the gemba; the place of real learning. For content modelling, to increase the likelihood of a model’s survival with real content, modellers must walk the gemba; to fish where the fishes are.
At Toyota, the gemba was the shop floor. Toyota software developers spent their time walking the gemba, learning how to build the shop floor systems that made their customers more proficient in their work.
Today, customers are online and engage with content through multiple digital channels such as websites, ebooks, mobiles, fridges, cars, and so on. That means we need to take into consideration both virtual (online) and physical (offline) gembas.
But, old habits are hard to break. In the same way, print folks hold onto their print media for comfort, digital folks do the same with the web. It’s so easy for us to focus almost entirely on wireframes and existing websites to find those pesky content types that make up our content models. However, to do so, is both short-sighted and negligent of us. They are many other places for us seek out those content types and evolve the content model. These include, product brochures, content authors, existing CMS systems, third party integrations, sponsors and stakeholders, other project teams, and so on. All of these gembas need to be revisited to enrich the content model; continually testing and tweaking as it as we go.
Walking the gemba is important. It’s where we learn and validate our content assumptions. It’s where the model makes contact with the real content.
Summary
No model should survive first contact with real content. But that’s okay, if that happens early and often during content design. Modelling at the gemba invariably adds that pragmatic and real-world edge to your content model, ultimately making structured content more sustainable for the business over the long term.
Oh, where is the triple Like button for this? When can we convince dev that it’s OK, in fact, preferable, that the content model evolves during the project? And how can we convince the content people that it’s good to look at the model with real content and help the model evolve to get the intended UX – and for all platforms?
So well articulated, Cleve!
Hey Rahel, you’re spot on. Right there is the holy trinity for real user engagement and experience: content, ux and technology. We still have a ton of challenges with content and technology, but I feel that the known unknowns like author experience, content architecture, content modelling are being done and gaining traction within the community (you, Ann Rockley, Noz Urbina, Jeff Eaton, Joe Gollner, Rachel Lovinger, Mike Atherton, just to name a few…)
But in the UX camp I still feel there are so many more unknown unknowns, and that’s definitely because I’m closer to content and technology, than to UX. So that’s something I’m going to focus on next, because as you implied above, we need all three to make this work, everywhere.
Thanks Rahel for stopping by 🙂
I love how you have articulated this. I’ve been thinking a lot about content models and how they need to be more human. Love the concept of going beyond digital to embrace the entire experience. Thanks for sharing this.
The more I talk to people about content models, the more I find that people are already working with them in one form or another. I love that. Thanks for the shout out Steve.
Amen, Cleve, to your comments and Rahel’s!
The severely truncated nature of the catch-phase — “No Model Survives First Contact With Real Content!” — definitely collapses a lot of nuance, and I think the nuance you’re adding is really important. The actual quote that I based it on actually reads:
“No plan of operations extends *with certainty* beyond the first encounter with the enemy’s main strength.”
The idea isn’t that everything falls apart once “reality” is encountered, but that any good strategist knows that the whiteboard perfection of a plan has to adapt to account for reality once it’s encountered. One of von Moltke’s *other* famous quotes hits on that idea: “Strategy is a series of expedients.”
To be effective, a content model will evolve in response to editorial and governance constraints, the availability of resources, the strengths and weaknesses of the implementation platform, and so on. Keeping the endgame in mind during the give and take negotiation — the series of expedients — is an essential part of a strategist’s work.
Nods, then smiles. Yes, I concur Jeff.
[…] Cleve Gibbon […]
[…] No model survives first contact with real content And now for a bit of theory. Content architect Cleve Gibbon talks about the limitations of content strategy. It always needs to be shipped before it can be improved via user feedback/interaction. ‘No model survives first contact with real content’. How true. […]
[…] afraid to roll up your sleeves and get started with intelligent content. To quote Cleve Gibbon: “No model survives first contact with real content … So go ahead and break your models, test them by running various scenarios using real […]
[…] how your content will look in a new environment. So when I read Cleve Gibbon’s article about how no model survives the first contact with real content it really took the pressure off. Cleve explains how content often breaks the design during the […]
[…] dies in der Literatur empfohlen und begründet wird (unter anderem vom Content Modelling Experten Cleve Gibbon: “no model survives first contact with real content”) und andererseits, weil wir als “Externe” zu wenig Einblick in die Ressourcen des […]
[…] as Cleve Gibbon states out “No model should survive first contact with real content. But that’s oka… It is important that you design your content model in a way it provides maximum flexibility to get […]